Sometimes I seem overly optimistic. Every new advance in technology always seems like a double edge sword to me. Plastic, the miracle substance. Throw away diapers. Cell phones (and texting), fossil fuel, nuclear energy. Every supposed advance has drawbacks, seemingly paradoxical It is almost as if we are not supposed to be doing this stuff to begin with. Maybe though, with this AI, we can learn and respond to some real intelligence for a change, rather than what we see today, trying to convince us they are intelligent as we go along with the programs of destroying our habitats. Maybe AI can evolve much faster than we in intelligence and thereby help save us poor rascals from extinction.
In the best case scenario, AI will make the least skilled 50% of humanity obsolete - the warehouse workers, truck drivers, and security guards. So what do we do with these surplus people whose upper limit of ability is being semi-intelligent muscle or standing in place waiting/looking for something? I don’t want to be stuck with the bill for financing their lives, and, left without limits, one can expect a faster rate of breeding in this cohort than the busy, high skill cohort. My money is on a lot of violence in the next decade or two.
AI is the current buzzword flavour of the year. Eventually it too shall fade away. Long ago buzz words were used in marketing spins to sell stuff to consumers:
• Automatic
• Turbo
• Electronic
• Integrated Circuit
• Computerised
• Robotic
• Green
• Eco Friendly
• Organic
Higher skilled labor is more likely to adapt than lower skilled labor and creative thinking is going to be augmented vs the wholesale replacement of unskilled physical labor. I think the trades are quite safe as well, as their physical labor is directed by their skill and knowledge. We plain need less meat and more brains. I don’t have an answer. Maybe to “do” is a call to state sponsored slaughter, which it isn’t. My point is that we will have a cohort of people who will be unable to support themselves and a high propensity to make more such people. What happens when you end up with 2 or 3 people needing social support to survive, during what should be peak wage earning years, per taxpayer? That seems like a Big Problem to me, one which we should get ahead of and do something about. We are all in a very sophisticated computer simulation so complex we think it's real, that we are real. I welcome our AI overlords and if you could just tweak the program a bit to make my life a bit more bearable I would be very grateful.
I should really be appealing to the 5th Dimension Organic Masters who push the buttons of the AI Overlords who keep our computer simulation running, despite them being blissfully unaware of their 5th dimension AI Overlords, who haven't got the foggiest about their 6th Dimension Organic Masters. Her Noodly Majesty laughs at any tangle of wires less complex than a Flying Spaghetti Monster... even if decent Pastafarians can’t spot Russell’s teapot, in which AI has stewed for decades.
I user to play with various AI tools to help people learn foreign languages. Neural machine translation (NMT) for example, speech recognition, image recognition - all of these can be employed to help people learn. Add crowd sourcing and you get a combination of human and machine intelligence, which offers many interesting benefits. For example, I took DeepL - a translating tool and first began by using as a translation aid. I soon discovered that you can use DeepL to improve imperfect texts, for example, which Germans have written in English. Neural MT sometimes changes the meaning of a sentence, so I let the crowd try to discover these errors and correct them. I’m currently developing the platform to include translation into simple language. This is something AI currently cannot do, but which is great when crowd sourced. The nice thing about all this is that the platform will be open source and free for anyone to use (If I don’t die first...).
In my opinion, we're just getting started with disruptive AI. It is transforming our lives and should be embraced. As a technical translator though, I have seen NMT engines like DeepL & Google Translate take over the job of translating to the point where I need to find other sources of income. As I work to integrate AI into language learning, I hope that my work in education offers me some security, at least until the machines finally take over that is. The dangers of a completely networked society are economic, political and cultural, but also moral on a fundamental level.
Firstly, the economic. There's a good book by two academic business advisors called 'Capitalism without Capital'. They argue that the increasing polarization of wealth in OECD countries since the 1980s and even right-wing populism can be accounted for by the digitalisation of the economy. The theory goes as follows. At the beginning of the last industrial revolution people worried that new technologies would mechanize labour to the point that human workers would be obsolete. It didn't happen. While stable hands were out of work, they could be re-employed as train drivers, engineers and so on. Indeed, while in terms of working and living conditions the first half of the 19th century was horrific, eventually a skilled manual workforce became so large that it organised. So grew the labour movement. Workers won rights, power shifted from capital to labour and the rural regions in Britain, America and areas of Europe that today are so impoverished thrived.
Fast forward to the next industrial revolution and these regions are de-industrialized as a new fibre-optic, global economic system develops. Again, there are fears that human labour will soon become obsolete, but again it doesn't. New jobs come into existence: software engineers, programmers, data analysts - but there is also there is a proliferation of work dependent on algorithmic networks such as media, marketing, research, the lifestyle and cultural industries. But there is a social problem with this new 'knowledge economy'. According to the book, it relies heavily on 'intangible assets'. Increasingly, business are investing in things that have no physical presence. Examples are software, apps, brands, patents, data, skills programmes, R&D, management systems, business models. Attempts have been made to measure the value of these assets and incorporate them into GDP, but not very successfully. In fact, as David Graeber argues in 'Bullshit Jobs', the capitalist system is now investing in human resources that serve no obvious point (though this is a bit of a separate argument).
Anyway, the corollary of all of this is the increasing concentration of wealth in cities where people with the skills and qualities needed to thrive in the knowledge economy congregate. Moreover, as intangible assets tend to compliment each other, collaborative networks of cognitive and cultural specialists form. Meanwhile, Hartlepool looks like the end of the world. Hence, a widening social disparity that is not just economic but cultural. Hence Brexit.
The other moral problem is more fundamental still - and it goes deeper than the surveillance and political abuses we all know so well. A networked world carries the threat of both authoritarianism and the destruction of the very foundations of what we understand as civilisation. For example, we are accustomed to recognise a distinction between a legal market of goods and services and a black market of criminal activity. But in a networked economy of virtual currencies and unaccountable transactions that distinction dissolves. Trafficked women are all over the porn networks; Mexican cartel money is held in western banks. There is no longer any coherent moral structure to the world.
While the right are abusing these new technologies in the worst ways, the left are not blameless. It is the generation of 68 revolutionaries who wanted a world of radical autonomy as much as the New Right. This is the result.
Kaku's take on the "future of humanity" is just scam to sell books (and it's also a pathology). I much prefer his technical stuff on Quantum Field Theory which I'm re-reading right now.
